Type 1 (T1DM) and Type 2 diabetes (T2DM) are characterized by loss of beta cell mass by apoptosis. However, the mechanisms leading to this beta cell loss are not fully understood. Recently, we identified thioredoxin-interacting protein (TXNIP) as a novel pro-apoptotic factor in beta cells and found that TXNIP expression is increased in islets of diabetic mice and in mouse and human islets exposed to high glucose. This suggested that TXNIP may represent a link between glucose toxicity and beta cell apoptosis and may play a critical role in the pathogenesis and progression of diabetes. TXNIP is a major regulator of the cellular redox state and induces oxidative stress by binding to and inhibiting thioredoxin. Beta cell-specific overexpression of thioredoxin prevented diabetes suggesting that inactivation of the thioredoxin inhibitor, TXNIP, may have similar protective effects. In fact, TXNIP-deficient HcB-19 mice, harboring a natural inactivating mutation of the TXNIP gene, as well as our beta cell-specific TXNIP knock out mice (bTKO), are hypoglycemic and hyperinsulinemic. Moreover, both HcB-19 and bTKO mice have increased pancreatic beta cell mass, further underlining the importance of TXNIP for beta cell biology. While the mechanisms by which TXNIP deficiency enhances beta cell mass are not known, our recent data suggest that IRS2/Akt/Foxo1 signaling may be involved. Interestingly, we also observed that the anti-diabetic drug, exenatide, reduces TXNIP expression and thereby decreases beta cell apoptosis. Therefore, we hypothesize that blocking TXNIP may help fight diabetes by enhancing endogenous beta cell mass. To test this hypothesis we propose the following Specific Aims: 1) To assess whether TXNIP deficiency can protect against diabetes, we will treat bTKO mice with multiple low- dose streptozotocin (STZ) as a model of T1DM, We will also cross bTKO mice with obese diabetic BTBR.ob mice as a model of T2DM and will determine whether beta cell TXNIP deficiency can rescue or improve the diabetic phenotype. 2) To investigate the effects of exenatide on TXNIP expression and beta cell survival in T1DM and T2DM, we will treat STZ-diabetic, NOD and BTBR.ob mice with exenatide and will assess islet TXNIP expression and apoptosis as compared to saline and glipizide treated animals. To determine the role of TXNIP, we will also treat bTKO mice with exenatide. 3) To study the mechanisms involved in TXNIP-mediated control of beta cell mass, we will analyze pancreata harvested under 1) and 2) using morphometry and TUNEL. We will also continue to investigate the role of Irs2/Akt/Foxo1 signaling using beta cell lines with TXNIP overexpression or with silenced TXNIP expression as well as primary bTKO islets. The results of these studies will provide new insight into the molecular mechanisms of beta cell survival and thereby the basis for novel approaches to the treatment of diabetes. In contrast to current therapies, these newer approaches will be geared towards preservation of the patient's own beta cell mass and insulin production, which should improve blood sugar control, quality of life and prevent disease progression.